Autobiographical memory (AM) plays a central role in people's construction of a self-concept, their experience of personhood, and the way in which they interact with world around them. AM is fundamental to a host of important behaviors, such as learning (e.g., Pillemer, 2003), communication (Bluck & Alea, 2002), goal pursuit (e.g., Singer & Salovey, 1993), and imaging possible future events (e.g., Szpunar, 2010). When AM is lost because of amnesia (Burgess & Shallice, 1996) or Alzheimer's disease (Bayles, Kaszniak, & Tomoeda, 1991) the consequences are profound. Similarly, quality of life can be severely reduced when AMs are intensely negative - many forms of psychopathology involve stress-provoking and intrusive memories, such as post-traumatic stress disorder (PTSD; e.g., Rubin, Berntsen, & Bohni, 2008). Since the origins of systematic empirical research on AM (Freud, 1915; Galton, 1883), the field has generated much knowledge about nature and function of AM as it develops across the lifespan (for review see Conway et al., 2010). A basic question that has been thoroughly investigated is how the content (e.g., specific details of events, narrative structure, etc.) and phenomenological characteristics (e.g., emotional intensity, vividness, etc.) of AMs change over time. In order to explain why AMs change over time, researchers often invoke repeated retrieval of previously experienced events (i.e. rehearsal) as a causal mechanism. However, the evidence for this claim is mostly correlational because AM studies usually measure retrieval retrospectively through subjective reports rather than manipulating it in prospective experiments. The broad goal of the proposed research is to test theoretical claims about how repeated retrieval affects AM and examine whether these effects differ in individuals with and without PTSD. More specifically, the proposed research will investigate how retrieval affects the content and phenomenological characteristics of AMs by performing prospective experiments in which the retrieval of AMs is manipulated as an independent variable. Critically, the theoretical claims that will be tested have direct implications for mental health. For example, one symptom of PTSD is the repeated retrieval of intensely negative, stress-provoking AMs - knowledge of how retrieval affects AM may enable the improvement of therapies involving the retrieval of memories to reduce the emotional intensity of trauma memories. The proposed research will advance Objectives #1 and #3 put forth in the NIMH Strategic Plan. The study will further our understanding of the behavioral and experiential factors that contribute to PTSD (Strategic Objective #1). In addition, the study will improve the efficacy of existing therapies tht are used for the treatment of PTSD, while also informing the treatment of other memory-related mental health disorders (Strategic Objective #3).